Low-voltage switch cabinet fault diagnosis method and system
By adopting multi-scale convolution and bidirectional timing feature extraction of dual-channel structures in the fault diagnosis of low-voltage switch cabinets, combined with attention-weighted fusion and residual enhancement, the problem of insufficient fault diagnosis accuracy and prediction capabilities in the prior art is solved, and efficient and accurate fault detection and prediction are achieved.
Patent Information
- Application Number
- CN202510216241.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-26
- Publication Date
- 2025-03-28
- Estimated Expiration
- 2045-02-26
AI Technical Summary
The existing low-voltage switch cabinet fault diagnosis technology faces the problems of reduced accuracy, high false alarm rate and lack of trend prediction capabilities, resulting in increased difficulty in fault detection and low equipment maintenance efficiency.
A low-voltage switch cabinet fault diagnosis method is adopted to realize fault diagnosis by performing multi-scale convolution and bidirectional timing feature extraction in a dual-channel structure, combining attention-weighted fusion and residual enhancement, and the fault prediction and response are achieved through dynamic time interval adjustment and continuous optimization of the prediction model.
It improves the accuracy of fault detection, reduces false alarm rate, has trend prediction function, improves the operation and maintenance efficiency and reliability of low-voltage switch cabinets, and reduces the downtime of equipment failures.